If you have much more short lived data than long lived data, tuning the bitcask merge settings might help. Note that more frequent merging also means more CPU and I/O usage.
I recently switched over to multi-backend exactly to deal with data in different buckets having different life spans. The major issue with that approach for us was that you cannot set which backend to use for a certain bucket from the client library, but have to do that from the riak console, which makes cluster setup less automated and more error prone.
_______________________________________________ riak-users mailing list riak-users@lists.basho.com http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
